There are no clinically recognizable mechanisms associated with AD progression. Recent studies of elderly normal (NL) subjects show that individuals with increased levels of plasma amyloid beta 1-40 (A40) are at higher risk for mild cognitive impairment (MCI) and Alzheimer's disease (AD). In transgenic AD (Tg) and wild type mice models, elevated plasma A40 is associated with decreased hippocampal (HIP) vasoreactivity (VR) and in Tg even prior to A deposits or brain damage. Our human pilot data in support of these animal findings suggest that plasma A40 is a biologically active vasoconstrictor of cerebral blood vessels, with pronounced early effects on HIP VR. However, it remains unclear if reduced VR is near the source of a clinical cascade that includes progressive structural brain damage, amyloid and tau pathology, and cognitive impairment. A vascular mechanism influencing progression is intriguing and consistent with the neuropathology and the epidemiology. However, since both plasma A40 and cardiovascular disease (CVD) impair brain endothelial and smooth muscle cell function, it is crucial to evaluate both factors and their interaction longitudinally. Our pilot data also show that A40 and CVD-risk independently target VR-CO2 affecting both overlapping and different cerebrovascular regions. The combined effects of elevated plasma A40 and CVD-risk on cognition, brain structure, and AD biomarkers remains unknown. We propose two NL aging studies. In Part 1 we will retrospectively establish the longitudinal relationships between plasma A40, CVD-risk, and cognitive decline in a large random community sample. We will assay plasma A40 and A42 levels in 1875 stored plasma samples from 625 randomly selected community residing NL elderly subjects, studied annually over a 3-year interval. In Part 2, we will conduct a prospective 2-year, three time point, longitudinal study of 200 NL individuals stratified by A40 level and CVD-risk. We will examine the relationships between A40 and CVD- risk factors as predictors of reduced VR-CO2 and AD related changes. All the required clinical, MRI, and biomarker measures were tested and validated during the funded cycle. This includes a new arterial spin labeling (ASL) method that precisely measures HIP and cortical perfusion and VR-CO2 without the spatial distortions typical in conventional echo-planar ASL. We will test four major hypotheses in cross-section and longitudinally: 1) elevations in plasma A40 levels preferentially reduce VR-CO2 in AD-vulnerable regions; 2) elevated plasma A40, in association with reduced VR-CO2, predict a cascade of clinical and biological changes related to AD; 3) elevated CVD-risk predicts reduced neocortical VR-CO2, and progressive deficits in verbal fluency and working memory; 4) for subjects with combined risks of CVD and high plasma A40, there is a synergistic decrease in VR-CO2 and increased cognitive and structural brain changes. This study has the potential to reveal an AD-related vascular mechanism associated with progression and to improve our understanding of the interactions between AD and CVD.